Can someone explain why we need inflation?

[nuclearhead]

I have read about inflation but I really don't understand why it's needed? Could someone explain?


The idea is that inflation tells you why the universe is so uniform right?

So let's assume the Big Bang starts with a singularity. From this emerges an infinite number of particles - bosons - which either become part of the background radiation or decay into matter-antimatter pairs. If the position of each particle after a Plank time say is entirely random, the probabilities of the positions of the particles in any finite area should follow a Poission distribution. OK, so the Universe the instant after the Big Bang is homogeneous and isotropic. So where does the clumpiness or variations curvature and temperature come from that inflation is needed to remove?

Does the clumpiness form shortly after the Big Bang and then have to be removed again by inflation?

They say how can the Universe be homogeneous if it has not had the chance to interact with other regions? But surely if everything started at a singularity then everything WAS in the same place and time at one point.

Does cosmic background radiation not follow a Poisson distribution?

Are they saying that if the Universe just expanding linearly, say, that it (the background radition) would clump together through gravity? Or that gravitational waves would make the Universe less flat?

 

[Matterwave]

Inflation was originally intended to solve 3 problems: The horizon problem, the flatness problem, and the monopole problem.

The Horizon problem is roughly what you're talking about here. It asks how can parts of the universe which are not in contact with each other, and couldn't have been in contact with each other, be uniform with each other? We look at the sky, and in opposite directions, the temperature of the CMBR is still roughly 2.7K, with inhomogeneities of the order one part in 10^5. The thinking is, even during the early epoch, these points of the sky were not in causal contact with each other, their Hubble horizons do not meet, so how could they have been in thermal equilibrium with each other? Inflation is supposed to account for this by basically making the Hubble horizon evolve in such a way that these points WERE in causal contact (before inflation) and then went out of causal contact afterwards.

Of course, you don't necessarily NEED inflation to explain this (or any of these 3 problems really). You can always appeal to initial conditions. The initial conditions of the universe were SUCH THAT it is roughly flat, or SUCH THAT it was homogenous even on scales larger than the causal horizon. But I guess many physicists don't like this "fine tuning" so to speak.

 

[nuclearhead]

Inflation was originally intended to solve 3 problems: The horizon problem, the flatness problem, and the monopole problem.

The Horizon problem is roughly what you're talking about here. It asks how can parts of the universe which are not in contact with each other, and couldn't have been in contact with each other, be uniform with each other? We look at the sky, and in opposite directions, the temperature of the CMBR is still roughly 2.7K, with inhomogeneities of the order one part in 10^5. The thinking is, even during the early epoch, these points of the sky were not in causal contact with each other, their Hubble horizons do not meet, so how could they have been in thermal equilibrium with each other? Inflation is supposed to account for this by basically making the Hubble horizon evolve in such a way that these points WERE in causal contact (before inflation) and then went out of causal contact afterwards.

Of course, you don't necessarily NEED inflation to explain this (or any of these 3 problems really). You can always appeal to initial conditions. The initial conditions of the universe were SUCH THAT it is roughly flat, or SUCH THAT it was homogenous even on scales larger than the causal horizon. But I guess many physicists don't like this "fine tuning" so to speak.

But I don't understand why they say the regions were not causally connected. They all started from the same point. The Big Bang singularity. Also any region in the Universe at any time one can detect background radiation from the big bang. So all regions are causally connected to the origins of this background radiation.

And if they weren't causally connected in the first place, how would gravity make the Universe all lumpy which inflation is there to solve?

Why would the Universe start in anything OTHER than thermal equilibrium?

What does that even mean to have two regions in thermal equilibrium with each other? Doesn't that just mean they're the same temperature? Or is that wrong?

If the Big Bang created an infinite Universe with an infinite number of random particles would they not be in thermal equilibrium?

A though experiment: If you divided a coffee cup into halves and let each side cool down on it's own. Even though there is no causal connection both halves are in thermal equilibrium with each other? How is that different?

I still don't get it. :(

 

[Matterwave]

The singularity itself is not well modeled by physics since it's a singularity. We can't really run physics backwards to t=0. So we can't make statements, using physics, such as "they were all in contact at one point in the singularity".

Any time after the big bang, there will be regions of space that are not causally connected to each other, because they are too far apart for light to have traveled from one region to the other given the time since the big bang. Remember that even though distances are much smaller back near the big bang, the time that light had to travel is also much shorter, leading to a small causal horizon.

Yet, we see that two regions which were effectively "never" in causal contact with each other still are at the same temperature (same temperature = thermal equilibrium). This is the horizon problem.

http://en.wikipedia.org/wiki/Horizon_problem

Again, this is not really a problem which necessarily NEEDS inflation. I emphasized in my last post that what you suggested, essentially initial conditions which are uniform, IS one possible explanation for the homogeneity. But physicists don't tend to like this "fine tuning".

 

[George Jones]

According to recent grad/research level texts, inflation, via quantum fluctuations, gives the most plausible mechanism for the the generation of perturbations:

The most serious of the above three problems is the horizon problem. As we have seen, there are possible solutions of the flatness and monopole problems that do not rely on inflation.

The most exciting aspect of the inflationary cosmological theories described in chapter 4 is that they provide a natural quantum mechanical mechanism for the origin of the cosmological fluctuations observed in the cosmic microwave background and in the large scale structure of matter, and that may in the future be observed in gravitational waves.

In the modern view, by far the most important function of inflation is to generate the primordial curvature perturbation ... It may generate other primordial perturbations too, including the isocurvature and tensor perturbations ... However, the historical motivation for inflation was rather different, and arose largely on more philosophical grounds concerning the question of whether the initial conditions required for the unperturbed Big Bang seem likely or not.

Originally inflationary scenarios were suggested as a pseudo-solution to certain pseudo-problems; these are only of historical interest today and the only reason to take the possibility of an inflationary phase in the early universe seriously is because it provides a mechanism for generation the initial perturbations.

 

[nuclearhead]

So it's not really needed?

 

[phinds]

But I don't understand why they say the regions were not causally connected. They all started from the same point. The Big Bang singularity.

The singularity was a point in TIME, not a point in space. All we know about the early universe is that it was staggeringly hot and staggeringly dense, but we do not know a size for it, except that it was NOT a point. It might have been infinite.

 

[nuclearhead]

The singularity was a point in TIME, not a point in space. All we know about the early universe is that it was staggeringly hot and staggeringly dense, but we do not know a size for it, except that it was NOT a point. It might have been infinite.

It was a point in time where the scale factor of the Universe is zero. Hence all matter is technically in the same point. Just like the pole of a globe has a greater circle with radius zero. And at the pole of a globe everything everything, West and East is at the same point.

Anyway, that's neither here nor there. If we assume an infinite number of bosons are created at the Big Bang with random momentums (because if we know a particles position exactly we can't know its momentum) then this is a good enough model for the Big Bang.

 

[Matterwave]

George's post is more up-to-date. It contains a more updated reason for evidence for inflation. See also the new Bicep 2 results.

 

[Lino]

... A though experiment: If you divided a coffee cup into halves and let each side cool down on it's own. Even though there is no causal connection both halves are in thermal equilibrium with each other? How is that different?

I still don't get it. :(

That's not a bad example. If you think of the complete coffee cup as pre-inflation and the two halves as post inflation, separating the two halves becomes the analogy of inflation. Both halves are in equilibrium and you can see why they must have been connected in the past. Imagine walking into a coffee shop at some point and randomly picking to half cups of coffee, chances are that one would be a fresh cup and one would be ... older(?) and so there would be a temperature difference / they wouldn't be in equilibrium. But if the two halves were the same temperature, you would think that the chances that the two halves were from the same cup would be much stronger.

Likewise, the two halves of the sky seem to be in equilibrium, so there is a good chance that the two halves were connected in the past ... and if they were connected in the past, something had to happen very very quickly to separate them (faster than the speed of light) ... and that something is what we call inflation.

Call it what you like, but something like inflation needs to have happened in the past to allow the standard model to make sense.

 

[Haelfix]

The money plot to understand why inflation (or something like inflation) is necessary is on page 17 and 18 of the following paper:
http://arxiv.org/abs/1404.2601

That is the picture behind the horizon problem.

Guth's original paper on inflation actually does a great job of explaining the other problems (monopole overabundance, flatness etc)

 

[nuclearhead]

The money plot to understand why inflation (or something like inflation) is necessary is on page 17 and 18 of the following paper:
http://arxiv.org/abs/1404.2601

That is the picture behind the horizon problem.

Guth's original paper on inflation actually does a great job of explaining the other problems (monopole overabundance, flatness etc)

Thanks! That was a good explanation. See if I got this right:

The "recombination phase" which protons and electrons bind together to form atoms, which also decouples the background radiation, would have produced random variations in temperature across space but when we look at the modern sky we see patches far apart that could not have been causally connected since this time have same temperature to 1 part in 10000.

So I think the source of my confusion is that I didn't realize the cosmic background radiation was formed at a post-big bang period, not at the Big Bang itself. (i.e. if the cosmic background photons had come from the singularity then there would be no need for inflation because it would have been causally connected).

Just to clear things up slightly. Could you explain why the recombination phase would have produced an uneven temperature distribution? What would this distribution be if not a Boltzman distribution? (I suppose random fluctuations just creep in over time?)

 

[bapowell]

Just to clear things up slightly. Could you explain why the recombination phase would have produced an uneven temperature distribution? What would this distribution be if not a Boltzman distribution? (I suppose random fluctuations just creep in over time?)

You mean what it would have looked like without inflation? If there isn't inflation, then we have no reason to expect uniformity of the CMB because we have no reason to expect that the baryon-photon plasma was even ever in thermal equilibrium.

 

[nuclearhead]

You mean what it would have looked like without inflation? If there isn't inflation, then we have no reason to expect uniformity of the CMB because we have no reason to expect that the baryon-photon plasma was even ever in thermal equilibrium.

Sure by surely we have no reason to expect non-uniformity either right? Unless there is a mechanism that produces non-uniformity? Which is what?

 

[bapowell]

Sure, you can say that causally separated parts of the universe give rise to CMB photons from a single Boltzmann distribution with some common, average temperature. But I would think you then need to explain this: it would be like different isolated societies on a prehistoric Earth that have never interacted all speaking the same language. Sure, that can just happen, but it's more likely that they would speak different languages. I don't think anyone would find that so compelling as to require an explanation for it.

 

[nuclearhead]

Sure, you can say that causally separated parts of the universe give rise to CMB photons from a single Boltzmann distribution with some common, average temperature. But I would think you then need to explain this: it would be like different isolated societies on a prehistoric Earth that have never interacted all speaking the same language. Sure, that can just happen, but it's more likely that they would speak different languages. I don't think anyone would find that so compelling as to require an explanation for it.

Well all human beings started off from a single source in Africa, allegedly, and through time they spread out and developed different languages. They didn't all start with separate languages.

So are you saying that from a single source, The Big Bang, different parts of the sky would naturally become clumped. (The only way clumping occurs is through gravity so I assume that gravity would clump things together which would get hotter?) So this clumpiness would occur between the Big Bang singularity and the age of recombination? If gravity is so weak and time until the age of recombination is so short, there couldn't really have been much clumping anyway could there? Well I suppose if everything is closer together the affect of gravity would be stronger. Is there any references somewhere to how this clumpiness occurs?

I would have imagined the the most likely of all possible universes is ones that started off without any structure? As structure implies high entropy. And as entropy is always increasing the initial condition must have been a state with lowest possible entropy.

How would the CMB look without inflation? Just more variation?

 

[bapowell]

Well all human beings started off from a single source in Africa, allegedly, and through time they spread out and developed different languages. They didn't all start with separate languages.

Yes, and wouldn't it be weird for all of them to independently develop the same language?? The scientific mind would strive for some kind of biological explanation for this.

So are you saying that from a single source, The Big Bang, different parts of the sky would naturally become clumped. (The only way clumping occurs is through gravity so I assume that gravity would clump things together which would get hotter?) So this clumpiness would occur between the Big Bang singularity and the age of recombination? If gravity is so weak and time until the age of recombination is so short, there couldn't really have been much clumping anyway could there? Well I suppose if everything is closer together the affect of gravity would be stronger. Is there any references somewhere to how this clumpiness occurs?

I'm saying that uniformity across acausal regions in the universe is not something one should expect from a causal process. Why are you assuming such initial differences should be small?

I would have imagined the the most likely of all possible universes is ones that started off without any structure? As structure implies high entropy. And as entropy is always increasing the initial condition must have been a state with lowest possible entropy.

Yes, this is about right. Look up "Boltzmann brains" if you aren't already familiar.

How would the CMB look without inflation? Just more variation?

I'd expect it not to be so remarkably uniform.

 

[Lino]

... So I think the source of my confusion is that I didn't realize the cosmic background radiation was formed at a post-big bang period ...

You probably know this, but just to point out, the CMBR is post bigbang and post inflation. The standard model tells us that inflation took the casually associated areas (including their equilibrium) and magnified them greatly. As a result the general equilibrium and the very very small fluctuations were magnified and are what are seen now in the CMBR in any direction.

As such the fluctuations are much less to do with gravity, then with the interactions / anniliations prior to recombination.

 

[nuclearhead]

Yes, and wouldn't it be weird for all of them to independently develop the same language?? The scientific mind would strive for some kind of biological explanation for this.

But if you take a hot cup of coffee and divide it into regions each region will independently cool to the same temperature.

I think the difference is that developing a language is increasing information which lowers entropy (which is compensated for by the increase in entropy coming from the sun burning).

Whereas the coffee cooling is increasing entropy.

Thus in order to explain why a big bang with low entropy hence highly non-structured would suddenly develop structure (i.e. increasing entropy) is something that needs explanation right?

The only thing that decreases entropy is gravity which causes clumpiness like stars and galaxys. Everything else like the sun shining increases entropy.

I would agree that from the evidence that if there is variation in the CMB and that large portions very far separated have similar temperatures then inflation is needed. I just don't get where the variation came from in the first place in such a short period of time.

I don't agree with the argument that because areas of the Universe look the same they must have been connected for more than an instant. (Take the coffee cup. Create it in an instant at the big bang. Then divide it into sections. All sections will have the same temperature over time.)

Sorry for keep going over it but it's the only way I'll fully understand.

 

[Lino]

... (Take the coffee cup. Create it in an instant at the big bang. Then divide it into sections. All sections will have the same temperature over time.)

Sorry for keep going over it but it's the only way I'll fully understand.

Also apologies from here nuclearhead, not sure what to add.

Maybe, think further on your analogy. I walk into a coffee shop and see lots of 'sections' of coffee floating around (different temperatures, different types of coffee, some sweetened, some with milk). I test a number of sections at random and they have the same equilibrium (temperature / coffee type / sugar / milk content), I'm surprised - maybe they come from the same original cup! If the sections were different / not in equilibrium, I'm much less surprised (after all, it's a big coffee shop).

Hopefully someone else can help you understand more clearly.

 

[nuclearhead]

Also apologies from here nuclearhead, not sure what to add.

Maybe, think further on your analogy. I walk into a coffee shop and see lots of 'sections' of coffee floating around (different temperatures, different types of coffee, some sweetened, some with milk). I test a number of sections at random and they have the same equilibrium (temperature / coffee type / sugar / milk content), I'm surprised - maybe they come from the same original cup! If the sections were different / not in equilibrium, I'm much less surprised (after all, it's a big coffee shop).

Hopefully someone else can help you understand more clearly.

Exactly, so if I discovered some milky coffee I would ask what MADE this coffee milky? Answer: Someone put milk in it. That is my question. What MADE all the variation in the first place? Other than that I would expect most coffee to taste pretty much the same... like coffee. And if they made all the coffee at the start of the day, I'd expect all the different cups of coffee to be at the same temperature.

It seems like people are assuming that there was a time when the plasma in the Universe was not in equilibrium and varied in temperature from place to place. Why is this an assumption?

 

[Lino]

... What MADE all the variation in the first place? ... It seems like people are assuming that there was a time when the plasma in the Universe was not in equilibrium and varied in temperature from place to place. Why is this an assumption?

Remember that after 14BY of expansion of the variations, we are talking about a variation of approx 10^-5 (not sure about the exact minus number). That's so incredibly small, and the same everywhere ... if it was coffee, it would all taste exactly the same and have to be made at exactly the same time, even though some is in New York and some is in HongKong!